During development, the epicardium contributes to the heart’s cellular components and guides and supports expansion of the coronary vascular network. A multitude of coordinated heparan sulfate proteoglycan (HSPG)-dependent pathways control epicardial cell activity. SULF1 and SULF2 – enzymes that modify HSPGs – present another level of control to refine signalling beyond ligand abundance. The regulation and specific roles of Sulfs in the epicardium are unexplored. Here, we used in situ hybridisation techniques and scRNA-seq to investigate Sulf expression over the course of embryonic mouse heart development. In addition, we carried out ATAC- and CUT&RUN-seq to identify epigenetic modification and potential transcription factor occupancy. Finally, we used ligand-receptor inference studies to explore HSPG-dependent signalling incoming to the E13.5 epicardium. Our study shows Sulf1 is expressed in epicardial progenitors and the forming epicardium, whilst Sulf2 is expressed more broadly throughout the myocardium during cardiac development. Sulf1 strongly co-localises with Wt1 in the epicardium, with levels of both diminishing to coincide with epicardial EMT and quiescence. Wt1 null embryonic hearts, which form an irregular epicardial layer, demonstrated lower levels of Sulf1 and upregulated Sulf2. We found that WT1 binds to an active enhancer within the Sulf1 transcription start site and a 5-bp mutation of the WT1 binding motif abolished enhancer activity. These data suggest a direct, positive regulation of Sulf1 transcription by WT1. Ligand-receptor inference analysis and subsequent functional assays identified HSPG-dependent signalling pathways impacted by this regulatory interaction. Namely, FGF-mediated proliferation and TGFβ-mediated EMT were found to be modulated. Our findings show WT1 regulates Sulf1 to refine HSPG-dependent signalling and drive processes such as proliferation and cell fate in the embryonic epicardium.

Background: Heterozygous truncating variants in the sarcomere protein titin (TTN) are the most common genetic cause of heart failure, a major cause of morbidity and mortality. This causality indicates that even two-fold changes in the amount of TTN can profoundly disturb cardiac physiology. Although a critical role of TTN in sarcomere formation and contractility is well established, the mechanisms regulating the TTN gene expression remain poorly understood. Therefore, we aimed to identify noncoding sequences that are critical for robust transcription of the TTN gene. Methods and Results: Using bioinformatics and assay for transposase-accessible chromatin with sequencing analyses, we identified evolutionarily conserved putative regulatory sequences in an intron 1 of the TTN gene. Homozygous deletion of the putative Ttn enhancer in mice led to embryonic lethality and heterozygous deletion resulted in allele-specific loss of Ttn expression. We performed reporter assays in human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) to demonstrate cardiomyocyte-specific gene expression driven by the putative enhancer. Conversely, deletion of this element in hiPSC-CMs downregulated TTN expression, impaired sarcomere development, and decreased cardiomyocyte contractility. Next, we performed modified massively parallel reporter assays in hiPSC-CMs to identify key transcription factors regulating the TTN enhancer and determined that disruption of NKX2-5 and MEF2 binding sites within the enhancer abolished its activity. We also generated transgenic mouse models to validate cardiac specific expression of the TTN enhancer and an essential role of NKX2-5 and MEF2 binding sites in its enhancer activity in vivo. Furthermore, whole genome sequencing analysis of 69 patients with unexplained dilated cardiomyopathy revealed one rare genetic variant within the conserved MEF2 binding motif. Introduction of the MEF2 site mutation in hiPSC-CMs led to decreased TTN expression. Conclusion: Discovery of a functional TTN enhancer provides new insights into the noncoding sequences that critically control expression of this key sarcomere gene. Manipulation of this element and molecules that bind it may advance therapeutic strategies to treat dilated cardiomyopathy caused by TTN haploinsufficiency.

The genetic control of heart development has been extensively studied, partly due to the prevalence of congenital heart disease (CHD). CHDs often arise from mutations in regulators of gene expression: altering gene dosage or changing spatial/temporal expression patterns of transcription factors. Gene expression is controlled in part by non-coding elements, however annotation and functional dissection of non-coding elements are lacking compared to the coding genome. Recently, our lab identified 8866 deeply conserved regulatory elements with shared accessibility and sequence conservation in human and zebrafish. Dubbed accessible conserved non-coding elements (aCNEs), these regions are enriched for cardiac enhancers and drive gene expression in spatially restricted compartments of the zebrafish heart. As over one-third of CHDs affect the outflow tract (OFT), we selected aCNEs that drove restricted OFT expression in zebrafish, examined the expression patterns of their orthologous human coordinates, and dissected their ability to act as OFT enhancers. Transcription factor motif analysis of aCNEs that drive expression restricted to the OFT reveal Fox motifs as overrepresented in these sequences. To test the hypothesis that Fox motifs restrict gene expression to the OFT, we removed these motifs in four separate OFT enhancers using site directed mutagenesis and show that gene expression spreads to other parts of the heart. Furthermore, addition of a Fox motif to a pan-cardiac enhancer is sufficient to restrict expression to the OFT. We are currently performing motif-swap and motif-affinity altering experiments to further dissect the regulatory code that governs OFT activity. While foxp4 is highly expressed in the zebrafish OFT by RNA sequencing and in situ hybridization, we show that OFT aCNEs are insensitive to foxp4 overexpression. We are currently screening other FOX transcription factors to understand the molecular control of OFT gene restriction. Additionally, we are generating foxp4 loss of function mutants to understand the role of this highly expressed transcription factor in OFT development. Altogether, this work has identified conserved novel cardiac enhancers active during OFT development and highlights the critical regions of these enhancers required for their function.

CHARGE syndrome is a birth defect associated with deletion or point mutations of the chromatin remodeller CHD7. Mouse models of Chd7 haploinsufficiency have severe cardiovascular defects, and conditional inactivation of Chd7 in the cardiopharyngeal mesoderm (CPM) recapitulates the cardiac phenotype of CHARGE patients. We investigated the role of CHD7 in the CPM and discovered genes and pathways affected in mutants at E9.5. We found that the balance of anterior and posterior second heart field (a/p SHF) progenitors is disturbed, with expression of aSHF markers reduced, pSHF markers increased, and early cardiomyocyte markers reduced in mutant embryos. We performed genome-wide profiling of CHD7 binding in cardiac progenitor cells, combined with transcriptomics from cKO embryos and identified direct CHD7 targets during the early stages of cardiac commitment and differentiation. Our analysis shows that CHD7 binds distal elements, with possible enhancer activity, regulating first and second heart field gene networks. Moreover, CHD7 shares a subset of its target sites with ISL1, including a well-characterised enhancer modulating Fgf10 expression in SHF progenitors vs. differentiating cardiomyocytes. CHD7 physically interacts with ISL1. We are now characterizing selected CHD7-bound putative enhancer elements, located >100kb from key cardiac TFs such as Isl1 and Hand1, using the enSERT assay. Preliminary data shows that 90% of the elements tested exhibit enhancer activity.

Atrial fibrillation and heart failure have a strong epidemiologic link, however the mechanisms underlying their mutual risk remain unresolved. We revealed a remarkable correlation of both gene expression and genomic features in the atria of atrial fibrillation (AF, Tbx5 deletion) and heart failure (HF, TAC banding) mouse models. Tbx5- and TAC-dependent gene expression in the left atria demonstrated a correlation coefficient of 0.8. Remarkably, virtually all dysregulated transcription factor (TF) genes in these two models demonstrated shared directionality, identifying 109 TF candidates for driving the shared gene expression response. Differential non-coding transcriptional profiling to identify altered cis-regulatory regions were exceptionally correlated between the models, including ~1,800 shared differentially expressed ncRNAs, of which ~400 emanated from accessible candidate regulatory elements. Integration with chromatin architecture allowed identification of lineage-specific shared AF and HF gene regulatory networks. A TBX5-driven cardiomyocyte-specific ncRNA-defined regulatory element drives expression of Klf15, an inhibitor of cardiac hypertrophy that is downregulated in both mouse and human heart failure. In contrast, a disease-specific ncRNA-defined regulatory element is upregulated with Sox9, an important transcriptional regulator of fibroblast activation. This work indicates that AF and HF initiate a shared genomic response in the atrium, including downregulation of a wild-type GRN that normally prevents injury-based disease signaling and upregulation of a disease GRN that promotes atrial remodeling, uncovering the common genomic underpinnings of distinct atrial diseases.